seismicpotentialofthemainactivefaultsinthe...
TRANSCRIPT
Seismic Potential of the Main Active Faults in the
Granada Basin (Southern Spain)
CARLOS SANZ DE GALDEANO,1 JOSE A. PELAEZ MONTILLA,2
and CARLOS LOPEZ CASADO3
Abstract—The main active faults of the Granada Basin are located in its central-eastern sector, where
the most important tectonic activity is concentrated, uplifting its eastern part and sinking the western
border. Several parameters related to the seismic potentiality of these active, or in some cases probably
active, faults in this basin are used for the first time. Many of these faults can generate earthquakes with
magnitudes larger than 6.0MW, although this is not the general case. The fault situated to the N of Sierra
Tejeda, probably the one responsible for the big earthquake of 25/12/1884, stands out, because it could
generate an earthquake with magnitude 6.9 MW. Although at present all the data needed are not fully
known, we consider that the final results show, as a whole, the average expected return periods of the faults
in the Granada Basin.
Key words: Granada Basin, active faults, seismic potential of faults, Betic Cordillera.
Introduction
The Granada Basin is located in the central sector of the Betic Cordillera (Fig. 1)
and is filled by upper Miocene, Pliocene and Quaternary sediments, covering the
contact between the Betic Internal and External zones. There exist many faults that
individualize the basin, also affecting its interior. In the basin and nearby there is a
noticeable, generally moderate to very moderate seismicity, although in its southern
sector occurred the so called ‘‘Andalusian Earthquake’’ (25/12/1884), Io MM = IX,
the most important one known in the region. Also, other earthquakes of remarkable
intensity occurred in this area.
1 Instituto Andaluz de Ciencias de la Tierra. CSIC, Universidad de Granada. Facultad de Ciencias.
Campus de Fuentenueva. 18071, Granada. E-mail: [email protected] Departamento de Fısica. Escuela Politecnica Superior. Universidad de Jaen. C/ Virgen de la Cabeza,
2. 23071, Jaen. E-mail: [email protected] Departamento de Fısica Teorica y del Cosmos. Facultad de Ciencias. Universidad de Granada.
Campus de Fuentenueva. 18071, Granada. E-mail: [email protected]
Corresponding author: Jose Antonio Pelaez Montilla, Departamento de Fısica, Escuela Politecnica
Superior. Universidad de Jaen, C/ Virgen de la Cabeza, 2. 23071 - Jaen (Spain).
E-mail: [email protected]
Pure appl. geophys. 160 (2003) 1537–15560033 – 4553/03/081537–20DOI 10.1007/s00024-003-2359-3
� Birkhauser Verlag, Basel, 2003
Pure and Applied Geophysics
In this paper we present the seismic potentiality of the active, or probably active
faults of the Granada Basin (these active faults are also presented here for the first
time, especially the map where they are represented). We start with the creation of a
database of faults in the Granada Basin, as was done in other works (WGNCEP,
1996; CDC-DMG, 1996). Initially, we consider the faults moving sensu lato during
the Neotectonic period, that is to say, during the last 10 Ma, although several of
them were older. Initially the number of faults considered was very large (507 faults),
but those shorter than 5 km have been eliminated, and among these remaining, we
only considered the faults with clear active features, or probably active. Here, a fault
is assumed to be active in a wide sense, that is to say, when it affects the Pleistocene
sediments. Nevertheless, in the Granada Basin, most of the faults considered active
have associated seismicity and other very recent geologic features, such as
geomorphologic scarps (see the notes of Table 1).
The database is focused on the knowledge of the seismic potentiality in the area
and will be useful as complementary information in the valuation of the seismic
Figure 1
Geologic setting of the Granada Basin in the Betic Cordillera, in southern Spain.
1538 Carlos Sanz de Galdeano et al. Pure appl. geophys.,
Table1
ThemostpotentiallydangerousfaultsintheGranadaBasin,accordingto
themaximummagnitudethatthey
cangenerateandto
theirreturn
periodofmagnitude
6.0M
W.Wefirstgivetheactivefaultsfollowed
bythoseconsidered
asprobablyactive,alwaysordered
accordingtotheestimatedvalueforthemaximummagnitude
thatthey
cangenerate.listhetotallongitude,pthedepth,dthedip,visthesliprateandtthereturn
period.
Faultname
andgeometry1
Segment
endpointN
andS(lon./lat.)
l(km)
p2(km)
dv3(mm/yr)
Maximum
magnitude4
MW,r M
w
t5(yr)to
MW=6.0
Nearby
towns
Notes
1.NofSierra
Tejeda,sl-n
Active
�4:184,36.964
�3:956,36.924
23.1
>10(g)
60�–90�0.125[10Ma]
6.7,0.2
a
>6.4,0.1
c
6.6,0.7a
6.9,0.3b
>6.3,0.5c
<1300(8)
Zafarraya,
Ventasde
Zafarraya
Evidenceofrecentmotion.
Possiblyresponsibleforthe
25/12/1884earthquake.
2.Granada,n
Active
�3:597,37.188
�3:521,37.083
16.8
>10(s)
60�
0.38[0.8Ma]
6.5,0.7a
6.6,0.2b
>6.3,0.5c
<510(1)
Granada,
HuetorVega,
Cajar,
Monachil
Thisfaulthas300mof
throwandaffectsthe
Pleistocenedeposits.It
movednoticeablysince
0.8Maago.
3.Padul,n
Active
�3:696,37.097
�3:576,37.003
15.2
>5(g)
50�–60�
0.16[5Ma]
0.35[1Ma]
6.4,0.6a
6.6,0.2b
>5.9,0.4c
<1200(5)
Durcal,
Padul,La
Malaha
Thisfaultisspectacularand
hasveryclearsignsofbeing
active.Itmovesjointlywith
thePadul-Durcalfault.
4.SantaFe,n
Active
�3:728,37.204
�3:647,37.107
13.0
>10(g)
60�
0.2[5Ma]
6.3,0.6a
6.5,0.2b
>6.1,0.4c
[<1200](6)
SantaFe,
Gabiala
Grande,
Alhendın,
Otura
Thisfaultisinferredfrom
itsgeomorphicfeaturesand
thedifferentnatureof
materialsatbothsides.This
faultanditsparallelhave
associatedanoticeable
seismicity.
5.Padul-
Durcal,n
Active
�3:638,37.049
�3:531,36.977
13.0
>5(g)
40�–60�
0.2[5Ma]
>0.35
[1Ma]
6.3,0.6a
6.5,0.2b
>5.9,0.4c
<1300(9)
Niguelas,
Durcal,
Padul
Thisfaultisspectacularand
hasveryclearsignsofbeing
active.Itmovesjointlywith
thePadulfault.
Vol. 160, 2003 Seismic Potential in the Granada Basin 1539
Table1
Continued
Faultname
andgeometry1
Segment
endpointN
andS(lon./lat.)
l(km)
p2(km)
dv3(mm/yr)
Maximum
magnitude4
MW,r M
w
t5(yr)to
MW=6.0
Nearby
towns
Notes
6.Atarfe,n
Active
�3:744,37.257
�3:662,37.193
10.3
>10(g
�s)
60�
0.15[5Ma]
6.2,0.6a
6.5,0.2b
>6.0,0.4c
<2100(10)Atarfe,Pinos
Puente,
Caparacena
Thisfaultpresentsan
importantscarpand
hassubstantialassociated
microseismicity.This
couldhelptoadjustbetter
thewidth.
7.ElFargue-
Jun,n
Active
�3:632,37.255
�3:546,37.179
11.7
>10(s)
60�
0.35[0.8Ma]
6.3,0.6a
6.4,0.2b
>6.1,0.4c
<790(2)
ElFargue,
Pulianas,
Pulianillas,
Peligros
Thisfaultpresentsaclear
activityduringthe
Quaternary.Ithas
associatedaveryclear
seismicity(serieofJune,
4th,1998).
8.Belicena-
Alhendın,n
Active
�3:703,37.188
�3:638,37.110
10.4
>5(sp)
60�
¿0.2?[5Ma]
6.2,0.6a
[6.4,0.2b]
>5.7,0.4c
[<3100](12)
Belicena,
Gabiala
Grande,
Alhendın
Itisknownfromseismic
prospectionandseemsto
haveidentifiedwithmany
earthquakes.
9.Albunuelas,
n
Active
�3:665,36.928
�3:561,36.916
9.8
>5(g)
40�–60�
0.14[5Ma]
6.2,0.6a
6.4,0.2b
>5.8,0.4c
<4200(13)
Pinosdel
Valle,
Albunuelas
ItaffectstheupperMiocene
andrecentQuaternary
deposits.Itis,therefore,
active.
10.Pinos
Puente,
n
Active
�3:757,37.265
�3:689,37.201
9.4
>10(s)
60�
0.4[5Ma]
6.1,0.6a
6.3,0.2b
>6.0,0.4c
<860(3)
PinosPuente,
Atarfe
Thisfaulthasagreat
seismicityassociatedwithit.
Itsthrowisabout2km.It
presentsimportantscarps.
11.Dılar,n
Active
�3:592,37.104
�3:533,37.048
8.3
>10(g)
60�
0.16[5Ma]
>0.33
[1Ma]
6.1,0.6a
6.3,0.2b
>5.9,0.4c
<1200(7)
Gojar,La
Zubia,Dılar
Thisfaultisclearlyactive
becauseitaffectsthe
Pleistocenedeposits.
1540 Carlos Sanz de Galdeano et al. Pure appl. geophys.,
12.Alitaje,n
Active
�3:774,37.265
�3:732,37.219
6.4
>5(sp)
40�–60�
0.1[5Ma]
5.9,0.5a
6.3,0.2b
>5.6,0.4c
<9000(14)PinosPuente,
Anzola,
PedroRuiz
Thisfaultisknownfrom
seismicprospection.It
affectsthePleistocene-
Quaternarydeposits,and
probablymovesjointlywith
thePinosPuentefault.
13.Obeilar-
PinosPuente,n
-sl
Active
�3:757,37.270
�3:845,37.257
7.9
>10(sp)g)
60�–90�
0.5[5Ma]
6.0,0.6a
6.2,0.2b
>5.9,0.4c
6.2,0.2
a
>5.9,0.1
c
<910(4)Zujaira,Casa
Nueva
Thisfaultisdeducedby
seismicprospectionand
isalsoobservableinthe
field.Itshowsrecent
motion.Itsthrowisof
theorderof2500mand
formsthenorthern
borderofaverysubsident
sector.Thehorizontal
displacementisnot
calculated.
14.PedroRuiz,
n
Active
�3:784,37.263
�3:751,37.216
5.9
>5(sp)
40�–60�
0.1[5Ma]
5.9,0.5a
6.2,0.2b
>5.5,0.4c
<9700(15)CasaNueva,
Anzola,
PedroRuiz
Thisfaultisknownfrom
seismicprospection.It
affectsthePleistocene-
Quaternarydeposits,and
probablymovesjointly
withthePinosPuente
fault.
15.Huenes,sl-
n
Active
�3:504,37.133
�3:529,37.101
5.0
>10(g)
60�–70�
0.25[0.4Ma]
5.9,0.2
a
>5.7,0.1
c
5.8,0.5a
6.0,0.2b
>5.7,0.4c
<2800(11)
Monachil
ItaffectsmiddlePleistocene
sedimentsandpresentsvery
recentandwellconserved
scarps.Thehorizontal
displacementisnot
calculated.
16.Escuzar,n
Probably
active
�3:665,37.040
�3:840,37.036
16.4
>5(g)
60�
>0.03
[10Ma]
6.5,0.7a
�6:8;0:3b
>5.9,0.4c
<13000(30)
Agron,
Escuzar,Pa-
dul
Thisfaultisprobably
active,althoughits
displacementisnot
important.
Vol. 160, 2003 Seismic Potential in the Granada Basin 1541
Table1
Continued
Faultname
andgeometry1
Segment
endpointN
andS(lon./lat.)
l(km)
p2(km)
dv3(mm/yr)
Maximum
magnitude4
MW,rMw
t5(yr)to
MW=6.0
Nearby
towns
Notes
17.Alcaucın,
n-sl
Probably
active
�4:101,36.918
�4:010,36.864
13.6
>10(g)
60�–90�0.175[10Ma]
6.3,0.6a
6.6,0.2b
>6.1,0.4c
6.4,0.2
a
>6.2,0.1
c
<1500(16)
Alcaucın,
Canillasde
Aceituno,
Sedella
Itdoesnotaffectmodern
material,butcontinuingto
theE,faultsofthistype
affectthePliocene-
Quaternary.Theyare
strike-slipandnormal
faults.Thehorizontal
motionisnotgivenbecause
appropriatereferencesare
notavailable.
18.Velezde
Benaudalla,n-
g
Probably
active
�3:445,36.898
�3:522,36.835
14.1
>3(g)
<30�
>0.15
[10Ma]
6.4,0.6a
�6.6,0.2b
>5.9,0.4c
<2900(20)
Velezde
Benaudalla,
Orgiva
Itmovedduringthe
neotectonicperiod,butwe
ignoreitsdepthandifitis
activetoday.Probablyitis
active,accordingtothe
geomorphologicindexes.
19.Sedella,n
Probably
active
�4:078,36.873
�3:978,36.846
10.8
>10(g)
60�–90�0.175[10Ma]
6.2,0.6a
6.5,0.2b
>6.0,0.4c
<1900(17)
Sedella,
Archez,
Canillasde
Aceituno
Itdoesnotaffectmodern
material,butcontinuingto
theE,faultsofthistype
affectthePliocene-
Quaternary.Theyare
strike-slipandnormal
faults.Thehorizontal
motionisnotgivenbecause
appropriatereferencesare
notavailable.
1542 Carlos Sanz de Galdeano et al. Pure appl. geophys.,
20.Guajar-
Fondon,n
Probably
active
�3:674,36.896
�3:595,36.834
9.8
>5(g)
60�
>0.12[5Ma]
6.2,0.6a
�6.5,0.2b
>5.7,0.4c
<5500(22)GuajarAlto,
Guajar-
Faraguit,
Guajar-
Fondon
ItaffectstheupperMiocene
deposits.Thisfaultis
clearlyneotectonic,butwe
don’tknowifitisactiveat
present.
21.Escoznar,n
Probably
active
�3:842,37.243
�3:788,37.167
9.7
>3(g)
60�
0.1[5Ma]
6.2,0.6a
6.5,0.2b
>5.5,0.4c
<11000(27)Escoznar,
Cijuela,
Lachar,
Romilla
Nueva
Thefaultcanberecognize
byitsmorphologyonly.It
musthaveaffected
Pliocene-Quaternary
materials.
22.NofSilleta,
n
Probably
active
�3:553,37.053
�3:641,37.050
8.2
>5(g)
60�
0.04[10Ma]
6.1,0.6a
6.5,0.2b
>5.6,0.4c
<20000(34)
Dılar
Thisfaultisprobably
active,althoughits
displacementisnot
important.
23.NofSierra
deLujar,sl-n
Probably
active
�3:422,36.897
�3:519,36.894
8.8
>10(g)
90�
>0.15
[10Ma]
6.2,0.2
a
>6.0,0.1
c
6.1,0.6a
6.4,0.2b
>5.9,0.4c
<2800(19)Izbor,Orgiva
Thisisasegmentofthe
faultthatextendstotheE
forseveralkilometersand
affectstheUpperMiocene
deposits.Thisfaultis
clearlyneotectonic,butwe
don’tknowifitisactiveat
present.
24.NWof
Jatar,n
Probably
active
�3:954,36.974
�3:894,36.912
9.1
>5(g)
60�
0.08[10Ma]
6.1,0.6a
6.4,0.2b
>5.7,0.4c
<8900(24)
Jatar
ItaffectstheupperMiocene
deposits,withoutevidence
ofrecentmotion.
25.Tocon-
Obeilar,n
Probably
active
�3:866,37.265
�3:963,37.250
8.8
>3(sp)
30�–60�
0.1[5Ma]
6.1,0.6a
6.4,0.2b
>5.5,0.4c
<10000(26)
Tocon,
Alomartes
Thisfaultisknownby
geophysics.Itaffectsthe
upperMioceneand
Pliocene-Quaternary
deposits.
26.Nof
Cubillas,n
Probably
active
�3:657,37.329
�3:748,37.305
8.6
>5(sp)
40�–60�
0.06[5Ma]
6.1,0.6a
6.4,0.2b
>5.7,0.4c
<11000(28)Cortijode
BerbeBajo,
Embalsede
Cubillas
Thisfaultisknownby
seismicprospection.It
affectsthePliocene-
Quaternarydeposits.
Vol. 160, 2003 Seismic Potential in the Granada Basin 1543
Table1
Continued
Faultname
andgeometry1
Segment
endpointN
andS(lon./lat.)
l(km)
p2(km)
dv3(mm/yr)
Maximum
magnitude4
MW,r M
w
t5(yr)to
MW=6.0
Nearby
towns
Notes6.Atarfe,nActive
27.NWborder
ofSierraArana,
n
Probably
active
�3:528,37.362
�3:556,37.307
6.7
>10(g)
60�
>0.12[5Ma]
5.9,0.6a
<6.3,0.2b
>5.8,0.4c
<4000(21)
Deifontes,
Cogollosde
laVega,
Iznalloz
Thisfaulthasbeenvery
activeduringthePliocene-
Pleistocene.
28.Zafarraya,
n-sl
Probably
active
�4:172,36.971
�4:098,36.960
7.0
>5(g)
60�–70�
0.1[5Ma]
6.0,0.6a
6.3,0.2b
>5.5,0.4c
6.1,0.2
a
>5.6,0.1
c
<9700(25)
Zafarraya,
Ventasde
Zafarraya
Thisfaulthasactedduring
theneotectonicperiod,but
thepresentevidencedoes
notlookimportant.
29.Daimuz
Bajo,n
Probably
active
�3:871,37.258
�3:830,37.204
7.0
>3(sp)
60�
0.08[5Ma]
6.0,0.6a
6.3,0.2b
>5.3,0.4c
<19000(33)DaimuzBajo,
Escoznar,
Lachar
Itisonlyaprobablefault.It
seemstohaveaffected
Quaternarymaterials.
30.Tablate,n-
sl
Probably
active
�3:508,36.960
�3:529,36.917
5.1
>5(g)
60�
>0.08[5Ma]
5.8,0.5a
<6.2,0.2b
>5.4,0.4c
5.9,0.2
a
>5.5,0.1
c
<16000(31)
Beznar
ItaffectstheupperMiocene
materials,butwedon’t
knowifitisactiveat
present.Consideringits
morphologicalfeatures,it
shouldbe.
31.Beznar-
Izbor,n
Probably
active
�3:546,36.927
�3:513,36.885
5.5
>5(g)
70�
0.08[5Ma]
5.8,0.5a
6.2,0.2b
>5.4,0.4c
<16000(32)
Izbor,
Beznar,Pinos
delValle
ItaffectstheupperMiocene
materials,butwedon’t
knowifitisactiveat
present.Probablyitis
active,accordingtothe
geomorphologicindexes.
32.Eastern
Cubillas,n
Probably
active
�3:669,37.324
�3:644,37.284
5.1
>3(sp)
50�
0.08[5Ma]
5.8,0.5a
6.2,0.2b
>5.3,0.4c
<23000(35)Embalsede
Cubillas,
Calicasas
Itisknownbyseismic
profiles.Itaffectsthe
Pliocene-Quaternary
materials.Possiblyit
continuestoact.
1544 Carlos Sanz de Galdeano et al. Pure appl. geophys.,
33.Wof
Cubillas,n
Probably
active
�3:720,37.300
�3:693,37.260
5.0
>4(sp)
60�
0.15[5Ma]
5.8,0.5a
6.1,0.2b
>5.3,0.4c
<11000(29)Caparacena,
Embalsede
Cubillas
Thisfaultisknownby
seismicprospection.It
affectsthePliocene-
Quaternarymaterials.
34.Canales,sl-
n
Probably
active
�3:442,37.172
�3:506,37.134
7.5
>10(g)
70�–90�
>0.08
[5Ma]
6.1,0.2
a
>5.9,0.1
c
6.0,0.6a
6.4,0.2b
>5.8,0.4
<6100(23)
Guejar
Sierra,Pinos
Genil
ItaffectstheupperMiocene
materialsandappearsto
havemodernmotion
accordingtoits
morphology.The
horizontalmotionisnot
estimated.
35.Lanjaron,sl
-n
Probably
active
�3:421,36.921
�3:495,36.919
6.6
>10(g)
90�
>0.3
[10Ma]
6.1,0.2
a
>5.8,0.1
c
5.9,0.6a
6.2,0.2b
>5.8,0.4c
<1900(18)
Lanjaron
ItaffectstheupperMiocene
materialsbutthereisno
directsignsthatitisactive
today,althoughsomeofits
geomorphologicindexesare
veryhigh.Thisfaultisa
segmentoftheonethat
continuestotheEthrough
theAlpujarrascorridor.
1(sl)-strike-slip,(g)-lowanglenormalfault,(n)-normal.Thefirsttypologyisthedominant
2(g)-bygeologicdata,(sp)-byseismicprospection,(s)-byassociatedseismicity
3Calculatedfromtheverticaldisplacementobservedinthelast[�Ma]
4(a)-usingtherelationship
MW=
MW(l)ofWELLSandCOPPERSMITH(1994),(b)-usingtherelationship
MW=
MW(l,v)ofANDERSONetal.(1996),(c)-
usingtherelationship
MW=
MW(A)ofWELLSandCOPPERSMITH(1994).Inboldface,actingasstrike-slipfaults
5Thenumberinparenthesisgivesthepositionintermsofhazardaccordingtothereturnperiodobtained.Activefaultsarefirstnumberedandsecondlythose
consideredasprobablyactive.Thecalculationofreturnperiodiscarriedoutassumingthattheyactasanormalfault
Vol. 160, 2003 Seismic Potential in the Granada Basin 1545
hazard. That allows us to open a new application of probabilistic methods to hazard
assessment, called probabilistic fault displacement hazard analysis (COPPERSMITH
and YOUNGS, 2000).
The obtained information must be gradually completed and corrected, because
the used data must be improved as the geologic and seismic knowledge of this area
increases. The trend and length of the fault, its geometry and the slip rate, have been
considered in the cases when it can be known from the accumulated throws. We used
empiric relations between the longitude and the surface of the fault and the greater
magnitude that it could generate. On the contrary, the existing paleoseismic
information in this basin is very scarce and consequently the paleoseismic intervals of
recurrence are not known, nor paleoseismic estimations of the displacement occurred
in earthquakes.
The aim of this paper is to estimate the seismic potentiality of the faults affecting
the Granada Basin, mind full of future applications such as the determination of the
seismic hazard.
The Network of Faults in the Granada Basin
This basin is affected by many faults distributed in three sets.
The faults of N70E to E-W direction form the first set. They are the longest and
oldest because many of them moved during the early and middle Miocene. Then they
moved as dextral strike slip, but from the late Miocene rested paralyzed or moved as
reverse and normal faults, according to its location. This occurred because from this
last time, the Betic Cordillera was affected by an approximateNNW-SSE compression,
combined with a near perpendicular extension, that in many points is more important
than the compression. At the same time, the region as a whole was rising.
The second set includes the NNE-SSW faults, which are very important,
especially to the east of the cordillera (e.g., faults of Lorca, Palomares and
Carboneras). They are present on the eastern border of the Granada Basin in its limit
with Sierra Nevada and further south. Its motion is normal, locally very important,
combined in some points with left lateral displacements (sometimes dominant). They
moved from the late Miocene to the present.
The approximate NW-SE faults form the third set. These faults are also present in
the eastern sector of the basin as well as in its interior, affecting areas such as Sierra
Elvira, Granada, Padul, etc. They basically move as normal faults, locally very
important and, as in the previous set, moved from the late Miocene to the present.
Some of them show a constant and noticeable microseismicity.
Moreover, there are low angle faults on the border of Sierra Nevada and more to
the South affecting the contact between the tectonic units of the Nevado-Filabride
and Alpujarride complexes (these complexes are the two lower ones of the Betic
Internal Zone). The displacement of these faults is in accordance with the direction of
1546 Carlos Sanz de Galdeano et al. Pure appl. geophys.,
extension (NE-SW), moving the hanging blocks to the SW; its present importance
generating earthquakes is not known, but seems to be negligible because there are no
earthquakes that could be associated with them.
These three set of faults that individualize the basin from the beginning of the
Pliocene are distributed in two systems: one is formed by the set of N70E to E-W
faults and the second by the other two sets. Owing to the stress orientation existing
from the late Miocene, the first system is not very active, as indicated by the geologic
features, and when it moves presents basically normal motions. Nevertheless, at
present this system can easily accumulate energy, and for this reason cannot be
discarded that it could move, even as reverse faults. Although both the focal
mechanism of the Andalusian earthquake (25/12/1884) and the exact position of the
hypocenter are not known, probably correspond to the movement of one of the E-W
faults of the South of the Granada Basin (the N of Sierra Tejeda fault).
The second system is very active and is responsible for many earthquakes
affecting the central and eastern sectors of the basin. The trend of the faults of its two
sets facilitates the extension, generally producing small to moderate earthquakes.
Figure 2 shows the total number of faults affecting the basin, presenting
displacements which occurred in the neotectonic period sensu lato. As can be
observed, some of them have an important length, particularly that of the N70E to
E-W strike. In the database they have been divided into their different segments;
Figure 2
Net of faults of the Granada Basin presenting movements during the Neotectonic sensu lato period.
Vol. 160, 2003 Seismic Potential in the Granada Basin 1547
otherwise in the calculation of the maximum magnitude that potentially could
generate, should result, unexpected high values.
The Database of Faults
To create the database we used the published cartography of the faults known in
the Granada Basin and its proximity, taken primarily from the geologic maps of the
region at 1:50000 scale, published by the Instituto Geologico y Minero de Espana
(IGME), although their trace has been corrected at many points with our own data.
Some of the faults have been known from seismic profiles made by Chevron�(RODRIGUEZ-FERNANDEZ and SANZ DE GALDEANO, 2001). The map obtained was
reduced to 1:100000 scale and digitized. For each fault we have considered the
following information.
Trace and total length of the faults. The location of each fault and the geometry of its
trace in the topography are needed in any evaluation of seismic hazard, as well as in
determinig different derived geometric parameters. The errors introduced during the
digitalization of the faults are usually less than 100 m and never exceed 150 m.
Given that generally there is no other type of reliable information, as could be
paleoseismic data, the total length of the fault obtained from its length trace is
considered as the maximum possible rupture of the fault. This is the value used in the
relationships between the length of the rupture and magnitude. Some authors
(WELLS and COPPERSMITH, 1994) consider that the rupture observed in surface
during an earthquake is only of the order of 75% of the length of the rupture which
occurred below the surface. According to this criterion, when using the length of a
fault as the maximum length of rupture, the seismic potentiality is possibly
underestimated. These and other authors (WYSS, 1979; ANDERSON et al., 1996;
PAVLIDES et al., 2000), established relationships between the observed dimension of
the fault (total length of the trace of the fault or rupture length in surface) and the
maximum magnitude. Using these relationships would avoid the underestimate of
the maximum magnitude that the studied fault is able to generate.
At first we had 507 faults able to generate earthquakes, potentialy active, using
the terminology of MACHETTE (2000). Their distribution is shown in Figure 2.
Using the lengths computed from the digitized traces, we have plotted the
cumulative number of faults N with length longer than a given value l (see Fig. 3).
Different authors (SCHOLZ and COWIE, 1990; WALSH et al., 1991; MARRETT and
ALLMENDINGER, 1992) have found, in some areas, that these two variables are related
through a relationship of the type
N / l�C:
1548 Carlos Sanz de Galdeano et al. Pure appl. geophys.,
In a log-log plot, such as in Figure 3, we should see, in case the faults under study
follow this kind of distribution, a straight line with slope given by the C parameter.
We can assume that this power law behavior agrees with the fault length distribution
in the Granada Basin and its surroundings, at least for the faults longer than 10 km,
approximately.
Concerning the study of the potentiality of the faults in the basin, we have
considered only those with a length greater than 5 km. This diminishes the number of
faults to be considered to 71.
Depth of the fault. In contrast to previous studies in which the lack of information
led to a consideration of most faults with equal depth (e.g. CDC-DMG, 1996), in this
work we try to establish, if not a value of the depth for every fault, at least an
estimation of it (a minimum value). In order to do so, we have used data of
associated seismicity, of seismic profiles and, when it has not been possible, simply
through several geologic features. We are aware that these estimations have a clear
margin of error, especially in the last case.
From this value, together with the length and dip, we estimated the fault-plane
surface, which was also used to estimate the maximum magnitude that the fault could
generate. The dip used was that detected on the surface, but numerous faults in the
Granada basin have listric geometry. This circumstance has been taken into account
in some cases.
According to MORALES et al. (1997) and GALINDO-ZALDIVAR et al. (1999), the
seismicity in the Granada Basin is concentrated at a depth of about 15 km in its
0.1 1.0 10.0 100.0l (km)
N
1
10
100
1000
C=3.6
Figure 3
Accumulative number of faults vs. length. A power-law fit for faults longer than 10 km has been computed.
Vol. 160, 2003 Seismic Potential in the Granada Basin 1549
northeastern part and towards the west until a depth of 20–25 km. This implies that
many of the faults of the basin reach these depths, approximately locating there the
main detachment level of the bottom of the basin.
Slip rate. To calculate this parameter, the throw obtained from the displacement of
several reference levels has been taken into account. The Tortonian marine
calcarenites, deposited approximately 8 Ma ago, are especially used because they
now reach very different heights. These range between 1830 m, in the western sector
of Sierra Nevada, and less than the present level of the sea in several sectors of the
Granada Basin: in the Cubillas and Pinos Puente areas (RODRIGUEZ-FERNANDEZ and
SANZ DE GALDEANO, 2001). The estimation of the throw of the faults that displaced
the reference levels and the interval of time in which it has been occurring gives
average values of the slip rate of the fault. In some cases the Pleistocene levels permit
obtainment of the slip, as it occurs in the Granada town and its surroundings. In
other cases some geologic features, such as the escarpment height, provide estimates
of slip rate of the fault blocks.
Activity of the faults. As previously indicated, the faults with a length greater than 5
km are only 71, all of them studied in detail. 37 of them have been discarded because
the known geologic and seismic data show that they have nonexistent, or practically
nonexistent, activity during the Quaternary. Among the discarted faults some of the
longest are included.
We have estimated only 15 faults of the Granada Basin as clearly active. These
faults show clear geologic features of recent activity and/or an associated seismicity.
Twenty other faults are considered as probably active. These 35 faults directly
considered are included in Table 1 and in Figure 4.
Maximum magnitude. To estimate the maximum magnitude that each fault is
capable of generating, and considering that generally there are no data of earthquake
magnitudes associated with a specific fault, we used different relationships proposed
by several authors between the maximum magnitude and the length or the surface of
the fault; additionally, the slip rate has been included in the estimation of the
maximum magnitude, as proposed in recent works. In this way we calculated several
values that enable us to verify the consistency of the results.
Independent of the value of the maximum magnitude obtained from the distinct
relations used, we estimated the error of the value obtained. For this, given that the
parameters appearing in the different proposed lineal relationships are accompanied
by their respective variance, simulations were performed using the Monte Carlo
method (RUBINSTEIN, 1981) when determining the variance (uncertainty) of the
estimated result.
1550 Carlos Sanz de Galdeano et al. Pure appl. geophys.,
The relationships used are the following: First, that proposed by WELLS and
COPPERSMITH (1994) between the moment magnitude MW and the length of the
surface rupture l. Also, we used the one proposed by these authors between the
moment magnitude and the rupture area A; this latter relationship is statistically
more robust than the former, in the sense that a greater number of earthquakes
was used to estimate the parameters and to establish the relationship, and as a
result the errors were slightly lower. Finally, we also used the relationship proposed
by ANDERSON et al. (1996) between the moment magnitude, the length of the
surface rupture and the slip rate v of the fault. With this relationship, since it
includes the fault-slip rate, we attempt to improve the fit between MW and l, thus
reducing somewhat the maximum magnitude expected for faults with high slip
rates.
Although the previous relationships are established for the moment magnitude,
as indicated by WELLS and COPPERSMITH (1994), there is no significant difference
between the magnitudes MW and MS in the range 5.7 to 8.0. This is precisely the
range of values of interest to us. Above the 8.0 value the MS scale saturates and we
have to work using MW instead. Below the 5.7 value the MS magnitude is
systematically smaller than MW. When expressing the maximum magnitude in
different scales for this area, the relationship proposed by LOPEZ CASADO et al. (2000)
can be used in order to relate the MS and mb scales for the Iberian Peninsula, in
addition to the cuadratic relationship proposed by the same authors between the mb
scale and the macroseismic intensity I0.
Return period. It would be desirable to determine the return period for a given
magnitude using the palaeoseismic information by studying the displacements
detected in the fault and taking into account the epochs on which these occurred,
however in this region such information is scarce.
To calculate the return period t, we used an approximation based on empirical
relationships. The expression used was
t ¼ d=v:
v is the slip rate, known for at least the main faults in the region. This is determined,as indicated previously, from the displacement detected in the fault during a given
time interval, although in some cases this may not be the present value of this
variable; this one could be known with geodetic measurements made during an
interval of years long enough to yield more confident results that those obtained from
geologic data. The variable d is the average coseismic slip, which again would be
useful to know from palaeoseismicity (movement that has generated a certain
earthquake in the fault). From the definition of seismic moment M0
M0 ¼ l � A � d;
Vol. 160, 2003 Seismic Potential in the Granada Basin 1551
where l is the rigidity modulus and A the surface rupture, we can calculate the
displacement d that caused an earthquake with seismic momentM0, or equivalently,
we can use the know relationship of HANKS and KANAMORI (1979)
MW ¼ 2
3logM0 � 10:7
to calculate the value of d that causes an earthquake of moment magnitude MW.
The slip rate, although barely influenzing on the maximum magnitude that a fault
may generate, considerably affects the return period of a given earthquake. The faster
a fault is, the lesser is the maximum magnitude that it can generate (ANDERSON et al.,
1996), given that it has less capacity to accumulate stress, and otherwise the return
period for a given earthquake is shorter because the average coseismic slip will be
already reached.
Results: The Distribution of the Active Faults in the Granada Basin;
Their Slip Rate and Their Seismic Potentiality
Table 1 shows the potentially more dangerous faults of the Granada Basin; those
considered active or at least probably active. The rest, particularly those included in the
list of the 71 longer than 5 km, couldmove, but in fact theymoved insignificantly during
the Quaternary, according to the geologic and tectonic features existing in each case.
The 35 faults considered have been classified according to themaximummagnitude
that they can generate. Similarly they have been classified in accordance with the return
period for themagnitude 6.0MW (I0 MM� IX, using the relationship of LOPEZ CASADOet al. (2000) for the region), considered as an indicator objective enough of their
potential risk. In Table 1 are initially firstly indicated the faults considered active,
arranged by the maximummagnitude that they could generate (we use the relationship
of ANDERSON et al. (1996) to obtain this classification, taking into account the slip rate
and the total length of each fault). Also, in the column where the return period for
magnitude 6.0MW is indicated, anordinal ismarked, indicating the relative importance
of the potential seismic of each fault in the basin (this is in relationwith the returnperiod
obtained in every case). The faults considered as probably active are situated
immediately after; equally arranged. In every case we indicate the typology, the total
length of the surface trace (l), the depth (p), the dip (d), the slip rate (v), the maximummagnitude that they can generate together with their uncertainty, the return period (t)
calculated/estimated for a 6.0 MW earthquake, and the villages nearest the trace.
Finally, notes and comments are added in each case.
The maximum magnitude to be expected is of the order of 6.9 MW in the case of
the so-called N of Sierra Tejeda fault, with a length of about 23 km and a slip rate of
0.125 mm/yr. This fault was probably the one which generated the Andalusian
earthquake (25/12/1884) (Fig. 5). Considering the length in the Granada Basin there
1552 Carlos Sanz de Galdeano et al. Pure appl. geophys.,
Figure4
SliprateofthefaultsoftheGranadaBasin.Thefaultsconsidered
active
aremarked
withthickcontinuouslinesandthoseprobablyactivewith
thickdashed
lines.Thenumberindicatestheirrelativeseismicpotential
(seeTable1).
Figure5
LocationofthemostenergeticearthquakeswhichoccurredintheGranada
Basininthelast600years,togetherwiththeactiveandprobablyactive
faults.TheearthquakesofAtarfein1431andArenasdelRey(Andalusian
earthquake)in1884standout.Theepicentersofthehistoricalearthquakes
andtheirmagnitudesareassignedfrommacroseismicdata.
Vol. 160, 2003 Seismic Potential in the Granada Basin 1553
are about thirty faults that could generate earthquakes with magnitudes higher than
6.0, although some of them have not been included in Table 1 because they do not
present clear displacements during the Quaternary.
The faults presenting higher slip rates are, fromNtoS:Obeilar - PinosPuente, Pinos
Puente, El Fargue - Jun, Granada, Belicena - Alhendın, Dılar, Padul y Padul - Durcal
and Lanjaron. These and others also important, although with lesser slip rate, are
shown in Fig. 4. Principally, the first ones are the faults presenting the shorter
recurrence periods. Among them the El Fargue - Jun and Granada faults stand out by
their slip rate. These two faults have produced importantmodern throws clearly visible
(they affect the Pleistocene sediments that are a good reference to calculate the vertical
slip). In both faults the age of 0.8Mahas been taken in a conservative estimation, as the
epochwhen the displacements began. Estimations of 0.6Ma or even 0.4Ma also can be
supported. Indeed, there is a group of faults, previously cited, whose slip rate can be
considered very similar, and moreover, that are very active from the seismic point of
view. This great activity can prevent the accumulation of energy required to produce
earthquakes with high magnitudes. Nevertheless, these faults concentrate most of the
important earthquakes of the Granada Basin (with the important exception of the
largest one, that of 1884). This fact can mean that they easily recharge the energy after
an earthquake and, consequently, that they continue moving at present as actively as
during the Quaternary.
The database permits the plotting of Figures 4 and 5 in which, for the first time,
the active and probably active faults of the Granada Basin are shown. The inspection
of these maps allows deduction of an interesting geologic feature: presently the more
active faults of the Granada Basin are located in its central-eastern part and
correspond to NW-SE faults. Beyond doubt, this sector concentrates the most
important tectonic activity, while during the late Miocene the faults with higher
motion were the more eastern ones. Probably in the future the same mechanism will
work again, and the higher activity will be translated to the west, likely in the area of
Agron (East of Alhama de Granada), where several recent earthquakes are located.
The movement of the present active faults uplifts as a whole, with some
exceptions, its eastern block, while sinking on the western side. This same mechanism
occurs in other sectors of the Betic Cordillera.
Conclusions
In this paper the estimation of diverse parameters related to the seismic potential
of the main faults of the Granada Basin is carried out for the first time. However,
estimations have some limitations, probably because the net of the faults is not
totally known in that some of them do not outcrop or their traces are hidden among
marly sediments and growings. Nevertheless, we think that the most important faults
are certainly shown in this work. Another limitation is that the geometry of the faults
1554 Carlos Sanz de Galdeano et al. Pure appl. geophys.,
is not well known, mainly as concerns the dip and depth. For this reason we have
adopted conservative estimations, avoiding exaggeration.
In any case, we consider that the final results accurately indicate the average return
periods to be expected in the Granada Basin, and additionally that they define clearly
several very active faults. From north to south are the faults of Obeilar - Pinos Puente,
Pinos Puente, El Fargue - Jun, Granada, Belicena - Alhendın, Dılar, Padul, Padul -
Durcal and Lanjaron. The return period marks an exaggerated seismic hazard in
several faults, according to their seismic data; this variable is known with more
uncertainty than themaximummagnitude expected. It is necessary to take into account
that the return period is inversely proportional to the slip rate and to the length of total
rupture. It is possible that in some cases the slip rate was overestimated. Also the length
of the faults may have been overestimated, in that the segmentation of faults is not
always well known. Moreover, it is highly unlikely that the faults move along their
entire length in every earthquake, especially in the longer ones.
Many of those faults can potentially generate earthquakes with magnitudes
higher than 6.0 MW, although this is not expected. In this sense the N of Sierra
Tejeda fault stands out because it could generate an earthquake of 6.9 MW, and
probably caused the earthquake of 25/12/1884.
Although this last fault presents the greater seismic potentiality in the basin, the
main active faults are located in its central-eastern sector, in the area where the
present tectonic activity is more important, uplifting the oriental block as a whole
and sinking the western one. Figure 5 points out the high correlation existing
between the more energetic seismicity of the area and the active faults with higher
seismic potentiality and higher rate of slip of the Granada Basin.
Acknowledgements
This article was funded by projects PB97-1267-C03-01 and REN2000-0777-C02-
01 RIES of the DGICYT and the group RNM 0217 of the Junta de Andalucıa.
REFERENCES
ANDERSON, J. G., WESNOUSKY, S. G., and STIRLING, M. W. (1996), Earthquake Size as a Function of Fault
Slip Rate, Bull. Seismol. Soc. Am. 86, 683–690.
CDC-DMG (California Department of Conservation - Division of Mines and Geology) (1996). California
fault parameters. CDC-DMG Open-File Report 96-08.
COPPERSMITH, K. J. and YOUNGS, R. R. (2000), Data Needs for Probabilistic Fault Displacement Hazard
Analysis, J. Geodynamics 29, 329–343.
GALINDO-ZALDIVAR, J., JABALOY, A., SERRANO, I., MORALES, J., GONZALEZ-LODEIRO F., and TORCAL, F.
(1999), Recent and Present-day Stresses in the Granada Basin (Betic Cordilleras): Example of a Late
Miocene-present-day Extensional Basin in a Convergent Plate Boundary, Tectonics 18, 686–702.
HANKS, T. C. and KANAMORI, H. (1979), A Moment-magnitude Scale, J. Geophys. Res. 84, 2348–2350.
Vol. 160, 2003 Seismic Potential in the Granada Basin 1555
IGME (Instituto Geologico y Minero de Espana). Geological Maps from Spain, Scale 1:50000. 1008
(Montefrıo), 1009 (Granada), 1010 (Lapeza), 1025 (Loja), 1026 (Granada), 1027 (Guejar-Sierra),
1040 (Zafarraya), 1041 (Durcal) y 1042 (Lanjaron).
LOPEZ CASADO, C., MOLINA, S., GINER, J., and DELGADO, J. (2000), Magnitude-intensity Relationships in
the Ibero-Moghrebian Region, Natural Hazards 22, 269–294.
MACHETTE, M. N. (2000), Active, Capable, and Potentially Active Faults - A Paleoseismic perspective, J.
Geodynamics 29, 387–392.
MARRETT, R. and ALLMENDINGER, R. W. (1992), Amount of Extension on ‘‘Small’’ Faults: An Example
from the Viking Graben, Geology 20, 47–50.
MORALES, J., SERRANO, I., VIDAL, F., and TORCAL, F. (1997), The Depth of the Earthquake Activity in the
Central Betics (Southern Spain), Geophys. Res. Lett. 24, 3289–3292.
PAVLIDES, S., CAPUTO, R., and CHATZIPETROS, A., Empirical relationships among earthquake magnitude,
surface ruptures and maximum displacement in the broader Aegean Region. In Proceedings of the Third
International Conference on the Geology of the Eastern Mediterranean, Nicosia 1998 (eds. Panayides, I.,
Xenophontos, C., and Malpas, J.) (Nicosia, 2000), pp. 1–10.
RODRIGUEZ-FERNANDEZ, J. and SANZ DE GALDEANO C., La cuenca de Granada. Interpretacion de su
estructura a partir de perfiles sısmicos, In La cuenca de Granada. Estructura, tectonica activa, sismicidad,
geomorfologıa y dataciones existentes (eds. Sanz de Galdeano, C., Pelaez, J. A., and Lopez Garrido, A.
C.) (Granada, 2001), pp. 109–116.
RUBINSTEIN, R. Y., Simulation and the Monte Carlo Method (John Wiley and Sons, New York 1981).
SCHOLZ, C. H. and COWIE, P. A. (1990), Determination of Total Strain from Faulting Using Slip
Measurements, Nature 346, 837–839.
WALSH, J. J., WATTERSON, J., and YIELDING, G. (1991), The Importance of Small-scale Faulting in Regional
Extension, Nature 351, 391–393.
WELLS, D.L. and COPPERSMITH, K.J. (1994), New Empirical Relationships among Magnitude, Rupture
Length, Rupture Width, Rupture Area, and Surface Displacement, Bull. Seismol. Soc. Am. 84, 974–1002.
WGNCEP (Working Group on Northern California Earthquake Potential) (1996). Database of Potential
Sources for Earthquakes Larger than Magnitude 6 in Northern California. USGS Open-File Report 96–
705.
WYSS, M. (1979), Estimating Maximum Expectable Magnitude of Earthquakes from Fault Dimensions,
Geology 7, 336–340.
(Received October 26, 2001, accepted February 26, 2002)
To access this journal online:
http://www.birkhauser.ch
1556 Carlos Sanz de Galdeano et al. Pure appl. geophys.,